A0615
Title: Functional principal component analysis for partially observed elliptical process
Authors: Yaeji Lim - Chung-Ang University (Korea, South) [presenting]
Abstract: The robust estimators of principal components are presented for partially observed functional data with heavy-tail behaviours, where sample trajectories are collected over individual-specific subinterval(s). The partially sampled trajectories are considered to be the filtered elliptical process by the missing indicator process, and implementing the robust functional principal component analysis under this framework is proposed. The proposed method is computationally efficient and straightforward by estimating the robust correlation function based on the pairwise covariance computation combined with M-estimation. The asymptotic consistency of the estimators is established under general conditions. The superior performance of our method in approximating the subspace of the data and reconstruction of full trajectories is demonstrated in simulation studies. Then the proposed method is applied to hourly monitored air pollutant data containing anomaly trajectories with random missing segments.
